Episode 38: Preston So, Oracle - Designing usable voice content

August 3, 2021

Preston So shares his thoughts and advice for new ways of delivering content. He offers examples and insights into omni-channel content audits, thinking about audiences holistically and techniques for enabling content strategy whatever the channel and format.

Available for download:

About this week's guest

Preston So

Preston So (he/him) is a product architect and strategist, digital experience futurist, innovation lead, developer advocate, three-time SXSW speaker, and author of Voice Content and Usability (A Book Apart, 2021), Gatsby: The Definitive Guide (O'Reilly, 2021), and Decoupled Drupal in Practice (Apress, 2018). 

A product leader at Oracle, Preston has led product, design, engineering, and innovation teams since 2015 at Acquia, Time Inc., and Gatsby. Preston is an editor at A List Apart, a columnist at CMSWire, and a contributor to Smashing Magazine and has delivered keynotes around the world in three languages. He is based in New York City, where he can often be found immersing himself in languages that are endangered or underserved.

He has been a programmer since 1999, a web developer and designer since 2001, a creative professional since 2004, a CMS architect since 2007, and a voice designer since 2016.


Episode transcript

Kristina Halvorson:

Hello and welcome to the Content Strategy Podcast. I’m your host, Kristina Halvorson, and every episode of this podcast I chat to established leaders and exciting new voices exploring our ever-evolving field of content strategy. We cover all the topics that inform how we shape digital content. From user experience design to customer experience, accessibility to content design and everything in between.

This week I am delighted to welcome Preston So to the podcast. Preston is a product architect and strategist, digital experience futurist, innovation lead, developer advocate, three times South by Southwest speaker, and author of Voice Content and Usability, Gatsby, the Definitive Guide, and Decoupled Drupal in Practice. A product leader at Oracle, Preston has led product, design, engineering, and innovation teams since 2015 at Aquia, Time Inc., and Gatsby. Preston, your list of accomplishments, and I'm not done yet.

Preston is an editor at A List Apart, a columnist at CMSWire, and a contributor to Smashing magazine. And he's delivered keynotes around the world in three languages. He's based in New York City where he can often be found immersing himself in languages that are endangered or underserved. Preston, welcome to the podcast.

Preston So:

Hey, Kristina. Thanks so much for having me. It's a real pleasure to be here on the Content Strategy Podcast today.

Kristina Halvorson:

It is really extraordinary the journey that you have been on in such a short period of time. Usually what I do at the top of the podcast is I ask people to share with me how they got to where they are now, and talk a little bit about how they have reached this focus on the importance of content and the role of content in designing experiences for our users. Can you talk about that just a little bit?

Preston So:

Sure. My career so far has been a bit of a winding road. It's been a very serpentine stretch of my life. I started out as a technologist, which means that I was really involved in computer programming. Back in 1999 is when I started actually writing in Visual Basic, my first programming language. I also did a lot of web design and web development back in the days when we were still using table based layouts, all sorts of really old techniques and 20 years ago was actually the time that I started working on websites for real.

I very quickly got an interest and a passion for the side of content that deals with architecture and implementation when it comes to how we actually enable digital content on the web, on mobile and all these devices. Of course, back in those times in the early 2000's, it was really primarily about the website.

It was still a relatively new thing, this whole transition from print to web. How do we take these things that are a broadsheet format, how do we take these things that are in a tabloid kind of layout, and bring them into a website in a way that makes sense? And a lot of the work that I focused on back in those days was around content architecture and content management. So all these software products that are out there, how do we actually make them work with all the different things that organizations need to use to make sure their content ends up in all the right places.

And it's really interesting because I've always had a very strong interest in user experience and design and content design, what we now think of as content design, but I didn't necessarily have a very deep understanding of a lot of these topics until I got really involved in some of the deeper implementations that I had the fortune to work on, in Time Inc. for example. I worked on Entertainment Weekly's website, which was a part of the Alexa Top 1000 back in those days. And one of the things I found very striking is, oftentimes a lot of the content approaches that people would take with the website would be, hey, let's just put all this stuff in a layout, plop it all up there and not think necessarily about how all of these different chunks of content, or all of these pieces of text and media should actually work together in a seamless way, in not just the website, but also in other types of media.

So in 2016, that's when I really started to work on some of these ideas around not only content strategy and content design as we think of it on the web, but also thinking about that other dimension of content strategy, which is, okay, you've got a great website that has great content. It's really well-structured, semantically valuable. You've got all these amazing attributes that you've got in your content. But how do we actually take that and make that something that's now amenable for things like voice interfaces or immersive experiences or some of these other user interfaces that are still coming up over the horizon.

Kristina Halvorson:

We have so much to unpack and I need four hours with you, Preston. Do you have four hours to give me? I'm going to start where you just ended. You've got this amazing website. Most of us do not have that amazing website. Most of us have websites like you said, "Yeah, let's make a layout, let's go ahead and design it, and let's throw a bunch of content up there and it will be good to go." That is as far as a lot of us have been, or a lot of our stakeholders, I'll say, have thought through the process. The internet, it's so big, we can publish as much as we want. Let's get this page up and then we'll go onto the next one.

At Brain Traffic we deal with a lot of very large organizations that have millions of pages, sometimes, of legacy content that they're trying to clean up in the first place. If someone is lucky enough to have the opportunity to break off a piece of a website, even several thousand pages and like, we're going to fix it so that it is usable, content is findable, and it's readable. What are the considerations that they need to take right from the outset so that we can get the content ready to be delivered and read or listened to or watched across all those different platforms? Where do you start?

Preston So:

Yeah, this is a really interesting question and I think it really depends a lot on the nature of the content and the way that a lot of these things have really evolved over the last few years. There's several axes to think about, and one of those obviously is the fact that, well, this transition that we're in the midst of right now, moving from written content to spoken content, to spatial content, these are things that really go well beyond. It's like the transition from 2D to 3D, right? It's really a very, very different kind of transition from the print medium to the web medium that we went through a couple of decades ago, and many people are still going through today.

So one of the things that I think is really important to think about, especially when it comes to these new experiences for content, the ways that we're delivering content that have really been never thought about, especially 20 years ago, is that a lot of our content is still very much written and very much oriented towards a written and textual medium and a visual medium. And if you think about it, a lot of the ways in which we transitioned over into websites with our content was just, okay, hey, we've got all of this content that's in these microfilm archives or in these very old databases that we need to move over into a website, but we're just going to forklift all of that over. And actually that's probably not the best approach I would say for many organizations, because you really do want to think about the value of that content on the web versus on a print format.

But one of the things that I think is really interesting about that transition from textual print content to textual web content is for the most part, a lot of it involved, okay, let's sprinkle in some links. Now we're going to have, let's say, some images interspersed throughout. We're going to make this web content, but that's not necessarily a huge distance or a huge leap from what we had before.

The difference, of course, today with a lot of the content that we're now delivering out to voice interfaces or to augmented reality interfaces or even smartwatches, is that we have to think about content very differently from the infinitely scrollable microfilm archive that our websites represent today. We don't have the luxury of a scroll bar. We don't have the luxury of clicking on links in a voice interface, for example.

And the really interesting thing is, you might be hearing me talk about all these very new-fangled novel approaches to delivering content, but the techniques for actually enabling a great content strategy remain largely the same. They're still very traditional. And that is, all of us know, at least those of us who really work hard to make sure the content that we work with is in a great place, is using the venerated content audit, right? We look for orphaned content. We look for content that's siloed over here. We make sure that all of our content is discoverable and legible. And when you think about that particular technique and that tactic, that's very much a procedural approach and a strategy that we can take when it comes to our other types of content as well.

One of the things that I often encourage all the organizations that I work with to do is, when you're looking at the ways in which your content is now going to manifest as mobile content, is now going to manifest as voice content, is now going to manifest as virtual reality content or digital signage content or some of these new approaches that really go very far away from what we have on the website today, is to take that same content auditing approach that you would normally undertake to make sure all of your content is in a good place and use what I call an omni-channel content audit, a channel agnostic content audit. To take a look at that full corpus of content, or even a subset of your content and say, okay, if I were to take this piece of content here and move this into a voice interface, move this into, let's say a virtual reality headset, how would that content behave? How would that content perform? Is it still as legible and discoverable as it was on the website? And a lot of the same ideas that we've used on the website, even if the kind of content that we're operating in is worlds away from this staid, well-established, traditional form of web content, those same ideas still very much apply.

I think a lot of people these days are very concerned about the advent of voice content, the advent of AR, VR content or digital signage content, because it completely overturns a lot of these paradigms that we're used to when it comes to designing for these user experiences or making them accessible. When in reality, I think from a content strategy perspective, a lot of that is very much still rooted in the same techniques and tactics that we've been using for many years now in the industry.

Kristina Halvorson:

I'll tell you right now that what most folks are thinking is, "This is so important, and this is so great. And I have exactly no idea where to start." Not with the audit, particularly because that's always a great, okay, well, let's dig in and see what's in there. And there are a lot of really fantastic methodologies and tools and ways in which we can get at that. But when we come back with those results, what needs to happen is we need to move forward together, aligned across departments, across roles, across business units and functions, around a set of guidelines and principles that say, and that inform, "Here's how we're going to be creating and revising content moving forward." And then there's all the governance, and the process, and the routines, and the rules, and accountability, and so on that have to be established.

So I have two questions for you. What are some of those baseline guidelines and principles that you tend to introduce to the organizations that you work with? And two, what advice or recommendations or tools do you have for people, for content strategists and content designers who are looking at the SEO team, and the brand team, and the content engineering team, and the tech writer team, and the product managers, and the executives to begin to say, "Okay, here's how we're going to pull folks together to get us marching in the right direction."

Preston So:

As you and I know very well, Kristina, a lot of times the biggest issue with content strategy, especially in such an interdisciplinary organization as those that we deal with, the hardest problem is usually the people, right? It's not really necessarily the technology or the design. And you touched on several aspects, I think, that are very, very crucial. And that is that none of us has the luxury to work completely in isolation, where we can drive all of these ideas for how the content should look or how it should operate on our own.

One of the things that I state in my book, Voice Content and Usability, is the very first thing, and I think this is something that a lot of folks have said over the years is, the very first thing that you need to agree on is, what is the problem and what is the actual need that you're trying to solve for?

I'll use an example that I use as an undercurrent throughout my book, which is the Ask Georgia Gov voice interface. Back in 2016, 2017 I had the unique opportunity to work with a team that worked on the very first ever voice interface for residents of the state of Georgia. We worked with the Digital Services Georgia team to establish a really interesting baseline for how we wanted to make sure that the content they'd already written that was amazingly versatile and amazingly well-written for the website, was something that we could actually transform into interfaces that could operate along the dimension of voice or along dimension of a written chatbot, for example. And there's so many different aspects of this that are very unique, of course, to the organizations that we work in.

And one example of this is that chances are that your organization doesn't look the same as another one and your editorial team doesn't look the same as another one either. You've got, obviously your technologists, you've got your product manager, you've got your usability team, you've probably got an accessibility team as well, focused on all these different aspects of how the content plays together. And the very first thing that we did, and I think this is something that I encourage everyone to do as well is, you always want to have some sense of discovery. You never want to just go willy nilly into a project like this, or any sort of implementation that involves delivering content to something new, even if it is a website, a separate website, without having a frank discussion about the needs and the requirements and the problem space that you're trying to actually solve for.

So we sat down for probably about a week or two with the Digital Services Georgia team, and just had basically a whiteboard discussion with no discussion of software, no discussion of architecture, where we wanted to really hash out what are the things that Ask Georgia Gov should solve? Because this is the thing that Digital Services Georgia came to us with. They said, hey, we've got a very good website that has some great frequently asked questions content, but we want to create a question and answer machine in a voice interface that can help make this a much easier experience for elderly Georgians that don't have, let's say the ability to use a computer as easily as those of us who are more digital natives, as well as those who are members of disabled communities in Georgia who might want to use a voice interface because that presents a better experience than a screen reader. So we already had some sense of the problem space in some sense of what we wanted to do. And that's just the first step.

With Ask Georgia Gov we knew that the demographics that we wanted to serve were very different from the sorts of demographics that we were going to actually look at on the website. And this is actually born out through some of the analytics and data that emerged.

The very first thing that we did was, hey, let's keep our scope very small. Let's focus on the question and answer content that we have on georgia.gov and not go bite off more than we can chew. And we also wanted to make sure that this was something that would be a proven use case and a proven set of requirements for those that we wanted to serve.

And so a lot of the work that we did initially was finding out more about not only those who are using georgia.gov from the standpoint of who these users are, what motivates them, what brings them to the website, but also those who are part of the Digital Services Georgia milieu of people and stakeholders, namely those who are picking up the phone and answering a lot of these phone calls from those who are calling these agencies directly and asking them about, how do I renew my driver's license? How do I register to vote? Because those are the kinds of people also that are very important to serving through a voice interface when it comes to really serving that audience in addition to the website.

I think that's one of the things that we often forget is, I think as technologists and as those who operate in terms of content strategy on the web, we often have a little bit of this tunnel vision, this little bit of this one dimensional perspective, which is absolutely valid because we work for hours and weeks on end on a website usually. But we forget oftentimes about those who might not actually have a computer at home, or might be using different forms of technology, who are actually accessing this content and accessing this information in ways that are completely devoid of any tether to the website that we operate in.

It's very important to think about, especially in these novel environments that go well beyond the confines of the jurisdiction of a normal website, how are some of the people that you might be seeing slip through the cracks actually looking for and experiencing this content? And it's not just a matter of talking to the SEO folks, for example. It's also a matter of talking to those who might be on the front lines of taking these calls, or helping some of these citizens or residents of your state government or your customers actually find this content.

Now, the second thing that we did, and I realize that was a very long answer for the first part, but the second thing that we did, and I'll keep this a little shorter, is during this process of content auditing, which I think is something that a lot of us now are very familiar with, thanks to of course, a lot of the hard work of content strategists who have come before us, is the notion of having a set notion of what those criteria are. What are the thresholds for recommendations that you want to make? You never want to go in with a very subjective attitude. But of course the problem with voice content, and the problem with some of these new forms of content 

One of the things I say in my book is, as opposed to a website, where oftentimes a lot of us now have got multiple websites under our belt. We know how content works on a website, for the most part, that we've worked with in the past. We know, generally speaking, how some of these content audits or some of these processes are going to go, is it's the very first time that a lot of us have ever tried to deploy content to a voice interface.

When it comes to a voice interface, however, a lot of these things might remain in flux. And I always recommend that people settle these criteria at the beginning with obviously a team of interdisciplinary stakeholders that represent all different aspects of the organization, but also to revisit those over the course of performing this audit and repeating that audit for certain sections of the content that you want to make sure are amenable for this voice interface, or what have you. Because chances are, you're going to discover or uncover pieces of content or sections of your texts that are not actually things that you expected to encounter in the initial settling of some of these criteria.

At the end, of course, you want to make sure to have at the very end, a full meeting with your stakeholders, with anyone who's involved in producing or managing or governance of this content, to have a frank discussion about some of these recommendations, because the fact of the matter is that even if you settle some criteria, even if you have some recommendations based on those criteria and one example of this is well, hey, if you have a call to action, get rid of that call to action, because you can't really do that very easily in a voice interface. Turn that into a more contextual link in the case of a website and make that something that's still legible and listenable in a voice interface.

Well, in certain cases that might not actually be a recommendation you want to follow because there's certain other aspects of value that you might not have considered. For example, let's say that you've got a brand manager who's saying, well, actually let's keep that call to action in there because we know for a fact that 65%, and I'm just making that number up, but 65% of our users who are looking at this content, according to our analytics, they're also smartphone users who are browsing the website on their phone, and if they're speaking with Alexa, they might also have a phone next to them. So they might be able to actually follow through with this call to action that says, learn more or read more at this website.

And it sucks because the answer typically, as all these things tend to be is, it depends, right? You don't necessarily want to say that you've got to get rid of all calls to action. You've got to get rid of all links. You've got to get rid of all these references to things that don't exist in a voice interface, because voice interfaces are very unique in that a lot of us oftentimes are fiddling around on our phones while we're talking with Alexa, or we might actually be speaking with Alexa or Google Home while we have a laptop or an iPad sitting right in front of us.

So these sorts of considerations really point to both of these parts, which really intersect that notion of knowing your audience, knowing your demographics, knowing who it is that you're trying to serve. And oftentimes, especially with these things that are off web or beyond the web, that requires you to really think about a much more holistic perspective of your audience than the typical audience you might be thinking of that's looking at your website or your content through a browser window.

Kristina Halvorson:

What's so interesting to me is I cannot help but think of the venerated case study of GOV.UK that emerged several years ago, where the government of the United Kingdom said, "You know what? We need to scrap the way that we have been doing our website content. We need to approach it from a completely different way." Which of course then was codified and documented in Sarah Richards, now Sarah Winter's book, Content Design. Which is we're going to use data, and we're going to use things that we know to better inform the content that we're prioritizing, how we're writing it, how we're shaping it, how we're prioritizing and presenting it, so that the people who need it the most get it when, where, and how they need it.

In that instance, it is citizens, right? And it's citizens with all the different needs. It is where you really do need to be thinking about at the core, usability, accessibility, inclusion. it sounds like just naturally that those things were woven into, of course, the process of the project that you're talking about with Ask Georgia.gov. The challenge I think is when your audience is everybody, it's really easy to rally around if we want people to get this information and be able to apply it and solve their problems or complete their tasks appropriately, these need to be some of our core principles. But when you're sitting in an organization where brand has a lot of power, how do you navigate that conversation when you're beginning to sit down and talk about voice and usability for the content that exists, when brand and marketing potentially have had such a heavy hand in creating it in the first place?

Preston So:

That's a really great question. And I think this really goes into some of the challenges around voice, as well, as a unique medium and as a unique channel for delivering content.

I think one of the things that we often forget is the brand voice and the marketing sense of how we want our content to sound or look on a page isn't necessarily the same as how it's going to sound when it's spoken. We can work as hard as we want to for hours and hours on end on honing that written brand voice, honing that perfect color palette or exactly the marketing assets that we want to make sure are represented throughout a website, but in a voice setting a lot of those things simply go out the window.

We've only had a few examples in the history of computing over the past 20 years where there's been such a clear personification or anthropomorphization or forging of identity in a disembodied user interface as we have in voice. And this is of course, Clippy notwithstanding, some of the personas that we used to have back in the day like Ask Jeeves, notwithstanding.

I think what's really interesting is that a lot of the ways in which we perceive websites, and this really goes into, I think, a lot of the psychology that's involved in some of the ways in which we perceive and consume content, well, a website is a website. If you think about a website and you think about the ways in which we characterize a website, or we think about how we brand a website or establish a website's identity, a lot of those are rooted in a visual context. We're playing in a visual sandbox when it comes to establishing that brand voice. And a lot of those things have to do with the same kinds of ideas and topics that we had to think about when designing for print.

But if you think about the fact that letting and kerning and typefaces and ligatures and color palettes and sans serif versus serif, a lot of those kinds of calculations have really remained largely unchanged in many regards from the print era. The ways in which we design newspapers are for the most part, apart from things like links, obviously, and nav bars and buttons and those sorts of things, they've been forklifted over to this environment where a lot of those things remain largely intact.

When you go to the voice environment however, these things completely fall apart. They completely desiccate. And I think it's really interesting that for example, one thing that Erika Hall says in Conversational Design, is the idea of conversation. Conversation is not a new interface. It's the oldest interface. If you think about the fact that only a mere 3,500, 4,000 years ago, that was the first development of cuneiform in the Mesopotamian civilizations that were so influential in codifying a lot of the notions that we have today of codes of law and irrigation and agriculture. That was the first surfacing of written literature and written texts. But we've been speaking as human beings, using this primordial habit, this primeval thing that we call human conversation and human speech for eons and eons and eons. I mean, people have no idea when we actually develop this ability to actually have a conversation like the one you and I are having right now, Kristina.

And this is a really, really tough thing, because when we have a conversation that's spoken, that gives us a very different kind of mentality and interaction with the brand than it does when we're actually having a written conversation involving correspondence or involving, let's say text. Because if you think about Alexa, if you think about Google Home, if you think about Siri, these are actually personified as people. And we don't normally think of things like GOV.UK, for example, or georgia.gov or white house.gov. We don't normally think of those as people, even though, of course, we understand in our heads that there's people behind this, there's 18F, there's people who are working on all of these aspects of what makes the website sing.

A voice interface however, because we hear them, we listen to them, they are necessarily embodied and sort of re-imagined as people. Now, this raises a whole bunch of interesting challenges, right? Because all of the work that our brand teams have done, all of the work that our marketing teams have done to really hone and really perfect the brand voice and the notion of how we want others to experience this content, really goes out the window. Because the fact is, when we work in a voice setting, all of that information is now being read to you by a disembodied voice that generally speaking, is the kind of person you're going to picture in your head.

And this is a question I ask at the beginning of chapter six of my book, which is, when you hear Alexa or Siri or Cortana or Google Home speak, who is the person you picture in your mind? Because at a very, very core level, at a very deep level, you do actually forge an image of a person in your head.

This really brings up, I think, a lot of the challenges that we now face around things like brand voice. And it's very ironic that we use that phrase, brand voice, because it is very relevant here. Because we now no longer have to think just about how this staid, abstract notion of text is splayed out across our browser view ports. We now have to think about, well, how do we establish authority and credibility and trust with the fact that there is this reconciliation that has to happen between somebody who has a completely different voice that is being represented by this voice interface and the person that we're dealing with?

Because one of the things that I think is very important to realize is that if you look at the vast majority and the overwhelming majority of voice interfaces up until today, and look at the historical underpinnings of how we've built voice interfaces in particular up until now, and only recently have we changed this, but up until recently, vast majority of voice interfaces were generally represented by a woman, a cis-gender, white woman with the general American dialect or a middle American dialect who speaks in a very, very certain way. And we hear this in Siri, we hear this in Alexa, we hear this at the airport, we hear this on the subway.

One of the things that I think is very interesting for those who are working on brand voice, and those who work in marketing now is, there is now this completely different dimension of considerations that we have to think about when it comes to inclusion and accessibility and equity, when it comes to delivering these voice interfaces. Not just the notion that now with voice interfaces we have a huge amount of ability to control the personality or the approaches that these voice interfaces take over, let's say screen readers, which are generally much more rooted in and tightly coupled to how that content was written in the first place.

Now that we have this spoken dimension of content, how do we deliver that in a way that is more amenable, as Chris Maury writes, for blind folks who need to be able to navigate this content in a way that's much more appropriate for blind users and not something that's rooted in a much more, let's say, heavy duty screen reader that's a lot tougher to manage.

By the same token however, I think it's very important to consider that a lot of the dialects, or a lot of the ways in which we speak as human beings are not represented in these voice interfaces. And what does that do to the notions of ... And it's not just a simple matter of something where, hey, this text is written in American English. We spell color without the U. We spell realize without the S, as the Britains do. It's much deeper than that.

Because if you're listening to a voice interface and you're constantly only hearing that voice that you hear everywhere, and that's the voice that is giving you information, how much trust do you have in that information as a member of a marginalized community or as a multiple-y marginalized individual. If you're somebody who has to code switch between let's say, straight passing and queer modes of speech, or you're a bilingual household and you code switch mid-sentence between English and Spanish, why are those toggles not represented within the context of a voice interface?

And so this is where a lot of these marketing teams and brand voice teams have to do a lot of thinking. Because the fact of the matter is that with these voice interfaces what's now emerging is a situation where you necessarily have a person that is standing in front of your voice interface and that is part of the picture that you're building with this voice interface.

I think it's very interesting to think about the fact that as we've seen a lot of the systems of oppression in the technology world continue to really manifest themselves in terms of automated racism, algorithmic oppression, a lot of those sorts of problems are now manifesting very clearly in voice interfaces as well. Because how much trust does, let's say, a Black indigenous and person of color actually place in a white woman's voice that is saying this information? Or how do we actually deal with this notion of misinformation that's coming out around COVID-19 or around some of the issues that we've seen with Facebook, when it's being presented in a voice that has a certain level of credibility in certain communities, but lacks that level of authority or credibility in certain other communities?

And one of the things I write in my book that I think is really relevant to this as well is, we have to think about the impact socially and economically of some of the work that we want to do with voice interfaces.

One example of this is, when you call an airline, when you call a hotel, when you call a call center, generally speaking, a lot of these call center staffers are people who speak English as a second language. There are people who are working in the global south, in the Philippines or in India or Indonesia, and they don't speak like we do. They don't speak the ways in which our newscasters do, in which our weather forecasters do, or how our Alexa devices sound.

There is some promising, really interesting changes in this direction, but the thing that I'm really, really worried about is the fact that a lot of times we're now fostering, in the same way that a lot of our problems in social media and technology have begun to foster this monoculture in our user experiences and in our content, we're now seeing that happen in voice interfaces as well. We're seeing this monoculture established, where the only accent that you can have is an American accent that is what we already hear everywhere, and really doesn't reflect the richness or the diversity or the multifaceted nature of how we as individuals speak within our own communities, how we code switch, how we do these toggles that really, to this day, are not represented anywhere in our voice interfaces.

And just to end this on an interesting note, just a few weeks ago I was on a podcast with my dear friends at QED42, a consultancy located in India, and we were talking about some of the facts of Indian English and how you look at some of the advertising some of the brands do in India, in Mumbai, for example, and they switch mid-sentence on these billboards or in these radio advertisements between English and Hindi. And how are those sorts of elements of how we use voice interfaces or how we interact with each other as humans reflected in voice interfaces.

One thing I can say right now is, I think a lot of people are saying voice interfaces are the future. It's going to be sort of this inflection point in computing. I think we're still a ways off from the point where we can really truly say that we've reached what Mark Curtis calls, the conversational singularity, which is that moment in time when the conversations that we have with voice interfaces are indistinguishable from the kind of conversation that you and I have right now, Kristina. And the question that we have to answer is, well, indistinguishable for whom?

Kristina Halvorson:

Preston, we are out of time. Thank you so much. I really appreciate you coming onto the podcast today. I wonder if you could tell people where they can find you and where they can pick up a copy of your latest book.

Preston So:

You can find my book, Voice Content and Usability at abookapart.com. It's now available. It's in print. And already have gotten some amazing feedback from readers about it. I want to thank everyone who's been so supportive so far.

I write about voice. I write about content architectures and content strategy at preston.so, my blog. But you can also find me at A List Apart as an editor and as a contributor. I write for CMSWire as a columnist. I've also written for Smashing Magazine.

You can find me on Twitter @prestonso, and on LinkedIn as well. And if you want to shoot me an email, you can do that at preston.so@oracle.com, or you can also use my contact form at my website, preston.so.

Kristina Halvorson:

Thanks a lot, Preston. Good news is I'm going to see you again very soon at Button, our content design conference. It's being produced by Brain Traffic in October, and you can find out more about Button at ButtonCONF.com. Thanks Preston, see you again soon.

Preston So:

Thanks so much for having me, Kristina.


Thanks so much for joining me for this week’s episode of the Content Strategy Podcast. Our podcast is brought to you by Brain Traffic, a content strategy services and events company. It’s produced by Robert Mills with editing from Bare Value. Our transcripts are from REV.com. You can find all kinds of episodes at contentstrategy.com and you can learn more about Brain Traffic at braintraffic.com. See you soon.

About the podcast

The Content Strategy Podcast is a show for people who care about content. Join host Kristina Halvorson and guests for a show dedicated to the practice (and occasional art form) of content strategy. Listen in as they discuss hot topics in digital content and share their expert insight on making content work. Brought to you by Brain Traffic, the world’s leading content strategy agency.

Follow @BrainTraffic and @halvorson on Twitter for new episode releases.